The Agent Skills Directory

[SAFE]: The skill utilizes the official @runcomfy/cli package from the NPM registry and communicates with vendor-owned domains (runcomfy.com, runcomfy.net). These are legitimate resources belonging to the skill author.
[SAFE]: Command execution is restricted to the runcomfy binary via the allowed-tools frontmatter, preventing arbitrary shell command execution.
[SAFE]: The skill processes external media URLs (images, audio, video) provided by the user. It proactively addresses security concerns by including a 'Security & Privacy' section that warns about potential indirect prompt injection from reference assets and instructs the agent to refuse harmful or non-consensual requests.
[SAFE]: Credential management is handled through standard CLI login procedures or environment variables, with documentation correctly identifying the local configuration path without attempting unauthorized access.

face-swap